Micro adjusting your DSLR

I've got my 'new' Sony a77ii and a whole bunch of lenses. Now comes the fun part - taking them all out and testing them in the field.

I enjoy testing cameras and lenses - always have. But I'm not really the 'brick wall and colour chart' kind of tester. I'm more of a 'how does it actually perform out in the real world' type of guy.

With one exception - the micro adjustment of lenses. For this sort of testing you kinda have to resort to some charts to be a little more 'exact' in your results. You are, after all, effecting the autofocusing results of said lenses.

Having said that, you don't have to spend hundreds of dollars on the fancy lens calibration charts that you need a degree in origami to be able to fold in exactly the right angle. A simple downloaded focusing chart off of the interwebs will do. Just do a google search and you'll come up with plenty of them.

But before we get underway with all of this AF micro-adjustment palaver, the question needs to be asked - "do I need to micro adjust my lenses?" And that's a very good question. The answer to which may very well be a resounding 'no'!

Are you happy with the results from your lenses? Do they look sharp to you? Do you have any fast (f1.4 - f2.8) lenses where autofocus will be critical? Or are all your lenses zooms, with an f4 - f5.6 range? If you don't have any fast glass, then chances are the aperture ranges you are using won't show up any 'slight' autofocus alignment issues.

But if you aren't happy with the sharpness you are getting out of your lenses, or if some seem sharp while others not so much, then maybe you need to consider in-camera micro-adjustments for certain lenses? That is, of course, if your camera actually supports such a thing. Many don't.

In-camera micro adjustment is a feature in more hi-end enthusiast or professional level DSLR's. You won't find this option on any of your Canon Rebels. And mirrorless cameras don't need to be micro-adjusted, since autofocusing happens directly at the sensor level, whereas DSLR's use separate AF modules. Sometimes these can be ever so slightly out of alignment, necessitating micro-adjustment. In the 'good-old-days', you would have to send you camera and lens(es) in to Canon/Nikon etc to have them calibrated if you thought the autofocus was 'out'. Photographers' talked of their camera either front or back focusing - especially portrait photographers who would lock the AF onto the eye of a model, only to find that the camera had focused ever so slightly in front of, or behind, the point of focus.

I haven't started using my a77ii yet, so I can't say that I'm having any focusing issues whatsoever. But I do have some very nice, fast, prime lenses for it. So I thought it was worth going through the motions, just for giggles. First up, the Minolta 85mm f1.4.

As mentioned earlier, their are lots of different suggestions online as to how to go about micro-adjusting your lenses. I decided to try a couple of different options, to see if they gave me the same results. Firstly, I found a chart where you set the camera up at a 45 degree angle and then focus on the chart laying flat, at a horizontal line with measurements down either side. Access your micro adjustment menu (check your manual since it will be different for each make of camera), and then shoot at a series of steps (I did 0,5,10,15,20,-5,-10-15-20) with your camera locked down on a tripod, 2 sec timer, aperture wide open, all stabilization turned off (since you are on a tripod). Take the images into Photoshop, enlarge to 100%, and check the areas of sharpness. It will soon become very evident as you go through the images whether this particular camera/lens combination is front or back focusing (or not).

It may be tricky to tell from the above re-sized image for the web, but the a77ii with 85mm f1.4 was ever-so-slightly front-focusing. On initial inspection a +10 adjustment looked the sharpest, but then I refined the process by shooting from +5 to +10 and reviewing the results. After inspecting this set closely, I determined that +7 was the sharpest of the lot.

Not totally convinced, I tried a second method. The 'flat-image-on-a-wall' test. This time you shoot some text on a wall with the camera parallel to the words. Same procedure with the camera (tripod, wide-open, self-timer etc). And wudda ya know - same result! It must be true 😀

Again, it might be hard to tell in these low-res images, but trust me - '0' was nowhere near sharp enough, and +7 sharpens things up nicely. Job done.

My a77ii can retain up to 30 lens adjustments - automatically. Meaning once you have adjusted the lens, the camera will remember it again once that particular lens is mounted on the camera. Brilliant. And I don't know about you, but I think even I might struggle to have 30 lenses to worry about?

So yes, it would seem that at least some of my lenses will benefit from micro-adjusting. And it also seems that either process will give you the exact same results.

But don't take my word for it. Download the charts, grab a fast prime, and try it yourself (as long as your camera supports micro adjustments). You may just breath some new life into that 50mm f1.8 that never quite felt sharp enough?

Micro Focus Adjusting a Canon 50D

I shot my first wedding in two years last Saturday - with my newly acquired 50D kit. It was a great day, they were a lovely couple, and I came away exhausted and exhilarated; with over 1000 images.

I processed the images through Aperture (yes, I still use Apple's RAW conversion software - but that's about to change), and was generally pretty happy with the results. I used both 50D bodies, and every lens in my arsenal on the day, so I felt that I gave my kit a really decent workout.

But on closer inspection, I began to feel that some of the images weren't as 'sharp' as I would have liked/expected. On an even closer inspection, I narrowed the images that I was less happy with down to 'my' 50D body and the 24-85mm lens. The 70-210mm lens on the same body was a tad sharper, but still not quite right.

My heart sank a little, since the images were indicating that the camera itself was faulty and the sensor not resolving sharply. I did some direct comparisons with my other 50D body a few days after the wedding, and these tests also showed me that on all my lenses, one body was producing sharper results than the other (all things being equal).

I considered sending the body away to Canon to check the sensor alignment - but since the camera is out of warranty I wasn't really thrilled about the cost involved. And then I remembered that the 50D has its own 'micro-adjustment' custom function (apparently the kind of adjustments that the Canon service centre would make anyway).

Since I wasn't happy with the body's performance, but didn't want to send it away to be 'aligned', I decided to give it a go myself.

There are numerous ways to 'micro' adjust your camera and lens combination, using either complex charts or moire pattern computer programmes. But I'm a much more practical, hands-on camera tester, preferring to go out and shoot real subjects rather than brick walls and test charts. So I found a 'real world' micro-adjustment test, and followed that.

Basically, you set the camera up with the lens set to its widest aperture (f2.8 / 3.5 etc) at its longest focal length (if it's a zoom). So for my 24-85mm lens I set it at 85mm @ f4.5. Then you find a target that has some detail and some 'roundness' to it, so you can gauge the movement of the focus point as you 'adjust' it. Many people use the limb of a tree. I chose our clothes line pole.

You then focus on the front of the subject and shoot a series of images, adjusting your alignment in 5 step increments. So, with the Canon 50D, I shot at -20, -15, -10, -5, 0, +5, +10, +15 and +20. That's 9 images in total. Shoot in RAW mode so that no sharpness is applied. Open them up in your RAW conversion software and save them all as high quality jpegs. Then open these in photoshop, zoom in 100%, and crop the final image making sure that you save them with an appropriate name so you remember what each setting was shot at. Go through and find the 'sharpest' number. For arguments sake, lets say it was -10. Subtract 3 from that (which gives you -13), reformat your card, take the camera outside on the tripod, and shoot again, but this time start at the number you just decided on (-13 in our example), and then shoot 7 images, going up by one step each time. This would give you seven images (-13, -12, -11, -10, -9, -8 and -7).

Go through the same process with these images; download them, convert to jpeg, enlarge 100%, crop and then compare the results. You may find that it gets very close at this stage, but a process of elimination should net you the sharpest result?

Do this with all your lenses, and the 50D will 'remember' each setting for each lens thereafter.

So did it work? You betcha. My results varied, but I was able to get the 'softer' 50D matching (and even surpassing) the sharper 50D body in all instances. My 24-85mm lens required a -6 adjustment, my 70-210mm required a +8 adjustment, my 10-22mm required a -8 adjustment, and my 50mm f1.8 required a +14 adjustment! As can be seen below, the 50mm adjustment was the most remarkable of all.

Canon 50D with 50mm f1.8 @ f1.8 with '0' adjustment.

Canon 50D with 50mm f1.8 @ f1.8 with +14 adjustment

It amazed me how 'unsharp' the 50mm was at its default setting before any adjustment. No wonder I wasn't very happy with the sharpness of some of the images I was getting from the 50D.

I've never had to do this with a camera before, as I've never been unhappy with the overall sharpness I was getting - and yes, I shoot wide open as often as I can, especially at weddings. But what about the other focal lengths (on the zooms) and the other apertures? Well obviously, because you are making these adjustments 'wide open', the results are even better when you stop down a bit (they always are) - so no problem there. And the other focal lengths? Well, it just seems that if you get the maximum length right then again, the others fall into place and are just as sharp.

If you are happy with the sharpness you're getting from you camera body/lens combination, then you don't really need to go through any of this. And as I said at the beginning, one body was fine for sharpness while the other was not. I've 'adjusted' the 'off' body now so that it falls in line with the body I was happy with - so for me the job is done.

But If you're not happy with the sharpness you're getting, it 'might' need a micro adjustment for your specific body/lens combination? Give it a try. You may be blown away by the results.